Generative AI explained by generative AI in everyday language

Prompt author: Malcolm, Content: Claude 3.5 Sonnet V2 via Monica AI. Image: Dall.E 3 via Monica

Human written context: Not everyone has jumped on the bandwagon yet, so here’s an explanation to help you get your head around what generative AI is and how it can help you in the nonprofit world.

AI inner monologue: hmm, explaining AI to someone who's never used it before... I should keep this super down-to-earth and relatable. Let me break this down into bite-sized chunks using everyday comparisons they'd understand from their life experience.

Hey there, nonprofit warrior! I hear you're curious about this whole generative AI thing. Let's break it down in a way that actually matters for your day-to-day chaos, shall we?

First up, imagine having a free (edit: or cheap) assistant who never sleeps, doesn't need health insurance (I know, budget constraints, right?), and can help you tackle that mountain of tasks you're juggling. That's kind of what these AI tools are like.

Picture this: It's 6 PM, you've got donor thank-you letters piling up, a board report due tomorrow, and your ancient CRM is acting up again. Instead of staying late (again), you could:

1. Ask an AI to draft those thank-you letters in minutes instead of hours

2. Have it turn your bullet points into a proper board report write-up

3. Use it to document the CRM issues and create a clear explanation for tech support

Sounds like magic, right? Well, it's not far off. These AI tools are like having different types of super-smart, always-available helpers. Let me break it down further:

Think of AI companies like car manufacturers, but instead of cars, they're making super-smart digital assistants. Here's how the AI landscape looks:

1. OpenAI (the Toyota of AI):

- They've got ChatGPT, which is like their top-of-the-line model. It's versatile and can handle almost anything you throw at it - from writing reports to summarizing data.

- Think of it as that efficient colleague who just gets things done - great for quick fixes and everyday tasks.

2. Anthropic (the Tesla of AI):

- Their model is called Claude. It's like a car with extra safety features - really good at being careful and explaining its thinking.

- Claude is more like that thorough consultant who explains everything - perfect when you need to understand complex problems or create detailed documentation.

3. Google (the GM of our AI world):

- They have Bard, which is good at pulling in recent info - helpful for grant research or staying up on sector trends.

The way these AIs work is kind of like having a super-smart pen pal who's read pretty much every book in the world. When you ask them something, they look through everything they've learned and piece together an answer - just like how you might give advice to a friend based on your life experience.

What makes them different from regular computer programs is that they can:

- Have actual conversations (like talking to a really knowledgeable friend)

- Write stories or articles (like having a writer on call 24/7)

- Help solve problems (like having a patient teacher who never gets tired)

- Create new ideas (like having a creative partner to bounce ideas off)

Now, here's the best part for our nonprofit friends drowning in work:

1. You can literally start using them in 5 minutes

2. They actually reduce your workload instead of adding to it

3. You don't need to learn complex features - just type what you need, like you're messaging a colleague

Let's look at a real-world example:

Say you've got that funding report due tomorrow, three volunteer orientation sessions to plan, and your CRM issues. You could:

1. Ask ChatGPT to draft the first version of your report from your bullet points

2. Have Claude help create a volunteer orientation script that actually makes sense

3. Use either to document the CRM issues and create a clear explanation for your tech support

The key thing is - these aren't another system you need to maintain. They're more like having a really helpful intern who's available at 11 PM when you're trying to finish that board presentation for tomorrow.

Look, I know trying new tools when you're already overwhelmed feels like adding another plate to juggle. But think of these more like putting down a few plates - they're designed to take work off your hands, not add to it. Start small - maybe just try using one for the next donor letter or board report. You might find it gives you back enough time to actually eat lunch at your desk instead of just staring at emails.

Remember, you're still the expert on your nonprofit and your community. These AI tools are just here to help you do your job better, not replace your judgment. Use them to save time on the boring stuff so you can focus on the important human-to-human work that really makes a difference.

So, what do you say? Ready to give your digital assistant a spin on that next task that's been hanging over your head? Trust me, it might just be the helping hand you've been looking for in your nonprofit juggling act!

P.S. (This part written by a human… ;)
This page of writing was done with 6 prompts:

Prompt 1: I need to explain to a 50 year old who hasn't used generative AI before what it is. the difference between different providers eg. anthropic and open ai. the difference between different models eg. gpt vs claude. I need to use metaphor and everyday language

Prompt 2: Now can you keep the tone and style the same but adapt for a specific role. An Ops manager in a nonprofit in North America who's time poor, struggling with lots of org challenges and technical debt, is feeling burnt out, there's too much to do and it's hard to learn a new tool they're not familiar with when they have no time. Can you explain how these tools might help them?

Prompt 3: I asked it to explain the tool I use called Monica AI. It went off on a tangent and made up useless stuff so I chucked that one out.

Prompt 4: The very first explanation you gave about car models needs to be updated with new models that are available can you check the internet for the latest AI models? (This is also gave me an annoying story I didn’t use)

Prompt 5: I want one longer story to help explain Gen AI. (This wasn’t helpful)

Prompt 6: Ok, I asked too much. Can you just combine the first two stories i asked for and not the concerns? (This did it!)

This whole process took about 5 mins. It would’ve taken me a looong time to write it like that. The amount of prompts is typical since I’m “talking” to software with “200” phd’s and the wisdom of a 3 year old.

The image was the 3rd attempt and still didn’t get the gist I was going for but it wasn’t worth spending time on.